An Introduction to MultiAgent Systems.

Model-based Bayesian Reinforcement Learning (BRL) provides a principled solution to dealing with the exploration-exploitation trade-off, but such methods typically assume a fully observable environments. The few Bayesian RL methods that are applicable in partially observable domains, such as the Bayes-Adaptive POMDP (BA-POMDP), scale poorly. To address this issue, we introduce the Factored BA-POMDP model (FBA-POMDP), a framework that is able to learn a compact model of the dynamics by exploiting the underlying structure of a POMDP. The FBA-POMDP framework casts the problem as a planning task, for which we adapt the Monte-Carlo Tree Search planning algorithm and develop a belief tracking method to approximate the joint posterior over the state and model variables. Our empirical results show that this method outperforms a number of BRL baselines and is able to learn efficiently when the factorization is known, as well as learn both the factorization and the model parameters simultaneously.

/pdf/bayesian-reinforcement-learning-in-factored-pomdps-xpra8a8w97.pdf

Bayesian Reinforcement Learning in Factored POMDPs

A scheme is developed for classifying the types of motion perceived by a humanlike robot. It is assumed that the robot receives visual images of the scene using a perspective system model. Equations, theorems, concepts, clues, etc., relating the objects, their positions, and their motion to their images on the focal plane are presented. >

http://ieeexplore.ieee.org/iel2/815/3148/00100651.pdf

Robot vision

Home PURPOSE OF THE CENTER: To develop the center to address state-of-the-art research, create innovating educational programs, and support technology transfers using commercially viable results to assist the Army Research Laboratory to develop the next generation Future Combat System in the telecommunications sector that assures prevention of perceived threats, and Non Line of Sight/Beyond Line of Sight lethal support.

How to… home page

The Object Management Group's (OMG) Model Driven Architecture (MDA) strategy envisages a world where models play a more direct role in software production, being amenable to manipulation and transformation by machine. Model Driven Engineering (MDE) is wider in scope than MDA. MDE combines process and analysis with architecture. This article sets out a framework for model driven engineering, which can be used as a point of reference for activity in this area. It proposes an organisation of the modelling 'space' and how to locate models in that space. It discusses different kinds of mappings between models. It explains why process and architecture are tightly connected. It discusses the importance and nature of tools. It identifies the need for defining families of languages and transformations, and for developing techniques for generating/configuring tools from such definitions. It concludes with a call to align metamodelling with formal language engineering techniques.

Model Driven Engineering

Reinforcement learning agents can successfully learn in a variety of difficult tasks. A fundamental problem is that they may learn slowly in complex environments, inspiring the development of speedup methods such as transfer learning. Transfer improves learning by reusing learned behaviors in similar tasks, usually via an inter-task mapping, which defines how a pair of tasks are related. This paper proposes a novel transfer learning technique to autonomously construct an inter-task mapping by using a novel combinations of sparse coding, sparse projection learning, and sparse pseudo-input gaussian processes. Experiments show successful transfer of information between two very different domains: the mountain car and the pole swing-up task. This paper empirically shows that the learned inter-task mapping can be used to successfully (1) improve the performance of a learned policy on a fixed number of samples, (2) reduce the learning times needed by the algorithms to converge to a policy on a fixed number of samples, and (3) converge faster to a near-optimal policy given a large amount of samples.

Multi-Agent Systems

A Process for Agent Societies Specification and Implementation (PASSI) is a step-by-step requirement-to-code methodology for designing and developing multi-agent societies, integrating design models and concepts from both object-oriented (OO) software engineering and artificial intelligence approaches using the UML notation. The models and phases of PASSI encompass representation of system requirements, social viewpoint, solution architecture, code production and reuse, and deployment configuration supporting mobility of agents. The methodology is illustrated by the well-known Bookstore case study.

/pdf/from-requirements-to-code-with-passi-methodology-2rlnl5ob6o.pdf

From Requirements to Code with PASSI Methodology

Holonic multiagent systems (HMAS) offer a promising software engineering approach for developing complex open software systems. However the process of building Multi-Agent Systems (MAS) and HMAS is mostly different from the process of building more traditional software systems as it introduces new design and development challenges. This paper introduces an agent-oriented software process for engineering complex systems called ASPECS. ASPECS is based on a holonic organisational metamodel and provides a step-by-step guide from requirements to code allowing the modelling of a system at different levels of details using a set of refinement methods. This paper details the entire ASPECS development process and provides a set of methodological guidelines for each process activity. A complete case study is also used to illustrate the design process and the associated notations. ASPECS uses UML as a modelling language. Because of the specific needs of agents and holonic organisational design, the UML semantics and notation are used as reference points, but they have been extended by introducing new specific profiles.

/pdf/aspecs-an-agent-oriented-software-process-for-engineering-35qdg637oe.pdf

ASPECS: an agent-oriented software process for engineering complex systems

Several agent-oriented methodologies have been proposed over the last few years. Unlike the object-oriented domain and unfortunately for designers, most of the time, each methodology has its own purposes and few standardization works have been done yet, limiting the impact of agent design on the industrial world. By studying three existing methodologies - ADELFE, Gaia and PASSI - and the concepts related to them, this paper tries to find a means to unify their meta-models. Comparing a certain number of features at the agent or system level (such as the agent structure, its society or organization, its interactions capacities or how agents may be implemented) has enabled us to draw up a first version of a unified meta-model proposed as a first step toward interoperability between agent-oriented methodologies.

A study of some multi-agent meta-models

http://www.agentgroup.unimore.it/Zambonelli/PDF/EAAI.pdf

Massimo Cossentino

Papers

Multi-Agent Systems

From Requirements to Code with PASSI Methodology

ASPECS: an agent-oriented software process for engineering complex systems

A study of some multi-agent meta-models

Process models for agent-based development